Interactive Films and Co-Construction

Interactive Films and Coconstruction
RENATO VERDUGO, MIGUEL NUSSBAUM, PABLO CORRO, PABLO NUNEZ,

and PAULA NAVARRETE,
Pontificia Universidad Catolica de Chile
Interactive Filmmaking is both an aesthetic and technological challenge. Steerable plots, where audiences are not passive
viewers but active participants of the narrative experience, require an engaging narrative model as well as a technologically
feasible structure. This article discusses the connection between aesthetics, cinema, and interactivity and presents a model for
interactive narration that is based on the audiences ability to read and interpret footage differently according to its context.
Through a detour narrative model it is possible to engage audiences in a coconstructive hypermedia experience while at the
same time minimizing the amount of footage required. An interface model that allows seamless hypervideo navigation through
graphic interaction is also discussed, and the interactive short film The Crime or Revenge of Fernando Moreno is presented,
along with user experience and usability studies that experimentally prove our hypothesis.
Categories and Subject Descriptors: H.5.1 [Information Interfaces and Presentation]: Multimedia Information Systems
Video; H.5.4 [Information Interfaces and Presentation]: Hypertext/HypermediaArchitectures, navigation, user issues
General Terms: Design, Experimentation, Human Factors
Additional Key Words and Phrases: Interactive film, coconstruction, storytelling, filmmaking
ACM Reference Format:
Verdugo, R., Nussbaum, M., Corro, P., Nunez,

P., and Navarrete, P. 2011. Interactive films and coconstruction. ACM Trans.
Multimedia Comput. Commun. Appl. 7, 4, Article 39 (November 2011), 24 pages.
DOI = 10.1145/2043612.2043617 http://doi.acm.org/10.1145/2043612.2043617
1.
INTRODUCTION
For every script ever written, many alternative stories have been left out. When commenting on a
film we often hear people say, What would have happened if. . . Immediately after, they imaginarily
rewrite the script, provide different endings, and change the course of events. Conventional films conceive audiences as passive receptors of a fixed stream of images, and watching a movie twice means
sitting through the same stream again. But what would happen if movies, instead of being static
repetitions, became aware of the audience and incorporated it into the unfolding of the plot? Stanley
Kubrick once said, If it can be written or thought, it can be filmed [Halliwell 1988], so why not explore
those stories that are being left untold, and let the audience decide what is to happen next? Exploring
The research was partly supported by the Center for Research on Educational Policy and Practice, Grant CIE01-CONICYT.
Authors addresses: R. Verdugo, Department of Computer Science, School of Engineering, Pontificia Universidad Catolica de
Chile; email: rverdugo@ing.puc.cl; M. Nussbaum, Department of Computer Science, School of Engineering, Pontificia Universidad Catolica de Chile; email: mn@ing.puc.cl; P. Corro, Institute of Aesthetics, School of Philosophy, Pontificia Universidad
Catolica de Chile; email: pcorro@uc.cl; P. Nunez,

Audiovisual Department, School of Communications, Pontificia Universidad
Catolica de Chile; email: pfnunez@uc.cl; P. Navarrete, Department of Computer Science, School of Engineering, Pontificia Universidad Catolica de Chile; email: pcnavarr@uc.cl.
Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided
that copies are not made or distributed for profit or commercial advantage and that copies show this notice on the first page
or initial screen of a display along with the full citation. Copyrights for components of this work owned by others than ACM
must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, to redistribute to
lists, or to use any component of this work in other works requires prior specific permission and/or a fee. Permissions may be
requested from Publications Dept., ACM, Inc., 2 Penn Plaza, Suite 701, New York, NY 10121-0701 USA, fax +1 (212) 869-0481,
or permissions@acm.org.
c 2011 ACM 1551-6857/2011/11-ART39 $10.00

DOI 10.1145/2043612.2043617 http://doi.acm.org/10.1145/2043612.2043617
ACM Transactions on Multimedia Computing, Communications and Applications, Vol. 7, No. 4, Article 39, Publication date: November 2011.
39
39:2
R. Verdugo et al.
Fig. 1. Issues regarding interactive films.
the narrative hidden behind a What would have happened if. . . is both a technological and artistic
challenge. Interactive filmmaking requires the creation of an aesthetically engaging experience along
with a technologically feasible structure, hence the double nature of the problem. To consider only
the aesthetic potential of interactive storytelling, without its technological aspects, results in a purely
theoretical discussion, while approaching the problem exclusively from a programmers point of view
creates a system that does not catch the audiences interest. Interactive audiovisual narratives must
respond to many challenges, of diverse natures including narration, interface, immersion, and technical implementation (Figure 1).
In this article we describe our approach towards Interactive Filmmaking. Our aim is to present
an authoring model to structure interactive films that considers audience participation as a pivotal
component of the narrative act, along with the interface model required to provide a seamless way
of navigation through the steerable plots of these audiovisual experiences. We begin by presenting
related work in the field, and then analyze the connection between aesthetics, cinema, and interactivity throughout history. We then discuss the narrative model that allows us to create authorial content
that, while still allowing authors to have narrative control, allows spectators to have an effect over the
unfolding of the plot. We also present the concept of coconstruction, which enables us to conceive the
story as an act of cocreation between filmmaker and viewer. We explain how coconstruction enables
the building of different stories, based on the viewers ability to read and reinterpret scenes in different
ways, according to interactively changeable contexts to those scenes, and explain why this is not only
an aesthetic need but also a technological and economical issue. Results of a narrative experiment
conducted with high school students are presented, to provide an empirical understanding of coconstruction. We then discuss our interface and presentation model, where hyperlink between one video
and the next is available during a specified time window, and the interaction occurs visually in the
form of dragging and dropping elements into or out of the film. We then conclude with the experimental interactive film The Crime or Revenge of Fernando Moreno, where our narrative model, the concept
of co-construction and our ideas regarding interface have been implemented. A survey that measured
the systems usability, the audiences reception of the film, and the assessment of co-construction as
a relevant part of the storytelling was conducted, and the results, along with an analysis, are also
presented in the final section of this paper.
2.
39:3
RELATED WORK
Interactive video has been thoroughly discussed and many different yet complementary definitions can
be found. From a general point of view, A video application is interactive if the user affects the flow
of the video and that influence, in turn, affects the users future choices [Stenzler and Eckert 1996],
while from a purely computational point of view, An interactive video is a digital video with hyperlink
type of interaction for browsing [Xu et al. 2003].
Regarding interactive video as a narrative tool, the broadest classification is the one presented by
Handler [2008]: all interactive movies fall into one of two quite different categories. One type is designed for a large theatre screen and is usually intended to be a group experience. The other type
is for a small screen and is viewed at home. It is a much more intimate experience, meant to be enjoyed by a single individual. An example of the first type of interactive film, designed as a collective
and democratic experience, is Terminal Time, by Michael Mateas [Domike et al. 2002]. Our approach,
which belongs in the second category of interactive films, is carried out through individual audiovisual
experiences, so that the user is the sole controller of the interactive event.
Attempts to use interactive video in narrative forms have had different levels of success, independently of whether they are designed as large or small screen experiences. Many of these systems rely on
complex hardware solutions (multiple screens, touch surfaces, etc.) [Tokuhisa et al. 2005; Knoller 2005;
Lew 2004; Atkinson 2008] rather than on theoretical models or structures that are flexible enough to
provide narrative freedom. Other works, like Interactive Drama [Szillas 2005], limit the users interaction to one character, and resemble a role-playing video game while others limit the interaction to one
user who performs in front of an audience [Marquez

et al. 2007].
The experimental hypermedia prototype HyperCafe [Sawhney et al. 1996] based its storytelling on
different conversations happening at the same time, providing users with the option to move from table
to table overhearing one conversation at a time while the others went on. This work also attempted
to provide a general framework for hypervideo. It described different types of link opportunities, and
provided different connections between videos, but was still aimed at one particular project, and not at
developing interactive films as a new audiovisual language. This approach, where the system acts as
a sequencer of previously edited chunks of video, is an example of a model for interactive storytelling
that provides narrative freedom while still keeping the system technologically feasible. As noted by
Brooks [1996] a storytelling system is not a magic box which creatively makes up a story when asked,
but a system of specially stored and organized narrative elements which the computer retrieves and
assembles according to some expressed form of narration. In these cases, the role of the computer is
then to match the desire of the audience (as expressed through an interface) to an appropriate selection
of content [Davenport 2002].
Regarding the narrative potential that interactive films have, one must not forget the impact that
interaction has over the viewers immersion in the story. This has been previously discussed in flow
principle in interactivity [Polaine 2005] where it is argued that interaction is the opposite of narration.
The cost of including interactivity in films cannot be the loss of flow [Csikszentmihalyi 1990]. The way
to include user participation within the interactive experience must be determined with this in mind,
and the users immersion must be considered as a key component of the overall experience. This issue
involves both the dramatic flow of the stories being told [Macfadyen et al. 2007] and the interface
through which users interact with the system [Johnson 2008].
3.
AESTHETICS, CINEMA, AND INTERACTIVITY
The understanding of interactivity as a relevant topic for artistic production can be first found in
the conscience of reception described by some aesthetic theories of the 20th century [Jauss 1982]. To
39:4
R. Verdugo et al.
Jauss [1982] and Vattimo [1997], for example, reception is ultimately the only truth of the artistic
event [Perniola 2001] and therefore it is impossible to consider a work of art or creation without its
encounter and interaction with an audience or spectator. Interactivity, before it had the potential to
become a pragmatic fusion between art and technology, was related with the interest of artists and
audiences to progress towards open works of art, where meaning and sense were not fully provided
but rather left to audiences interpretation and reading. This phenomenon has been studied and inventoried by Eco [1989] in his book The Open Work, where he describes texts as fields of meaning, open
for reinterpretation and contextualization. Dorfles [1984] presents the idea of interval as a space for
the structural understanding of an artistic creation through the reconstruction of the viewers own
conscience, perceptual memory and sense of time and space. These theories can all be understood as
early stages of audience involvement with works of art.
In cinema, understood as both a form of mass media and a technically based channel of artistic
expression, it is possible to find different levels of sympathetic relations between film and audience
that go from passive contemplation to ingenious technical and narrative operations that provide higher
levels of viewer immersion and interaction. A films relation and connection with its viewers derive
from the illusion of reality it provides, and the first linguistic device it has to relate with interactivity
is montage. This allows creators to play with the structure of the film, to choose what will be shown and
what will be hidden, to use the suggestive power of ellipsis and ultimately determine where viewers
will have to fill in the blanks and make up their own story.
Technical procedures opposite to montage may also provide forms of interactivity: temporal continuity and spatial continuity. A sequence shot (the uncut following of an action by a moving camera) as
used by Rosellini, De Sica, and Italian neorealism in general, was interpreted by Andre Bazin as a gesture of openness towards freedom of discernment, freedom of the spectators attention, and a form of
dynamic sensory realism. The same author noted that in Orson Welless films, the use of an extensive
depth of field (portion of a scene that appears in sharp focus) was an invitation so that viewers would
travel through the image, and freely relate the elements that coexist, whether they remain static or in
movement, within an extensive space.
Another way of interactive immersion is achieved in cinema through the use of the offscreen space,
and the cameras position. What is called the fourth wall in theatre can be used with interactive
purposes in cinema, as in Rashomon (1950) by Akira Kurosawa and Lady in the Lake (1947) by Robert
Montgomery. In Kurosawas film, a masterpiece considered as the maximum expression of offscreen
space, the cameras position identifies with the point of view of judges that listen to different versions
of a criminal event. Because of its open ending, where no particular version is revealed as the truth, the
identification between the judges and the audience occurs in the form of an endless discussion about
the final deliberation, and the determination of what really happened. In Montgomerys film all of the
action is viewed through the main characters eyes, and the movement of the character determines
the movement of the camera. The audience is invited to view and feel the story just like the character
does, and the action is followed only through immersion with him. In films like A Bout de Souffle
(1960) by Godard, and Funny Games (1997) by Michael Haneke, interaction was achieved by looking
into the camera, ignoring the fourth wall and talking directly to the viewer. Split-screen narratives,
like Timecode (2000) by Mike Figgis, and Pillow Book (1996) by Peter Greenaway, also provide ways of
interaction by the presentation of simultaneous actions. Interactivity is, in these cases, reduced to the
itinerary of the viewers attention, which must jump from one space to another.
These connections between film and audience are only the surface of the interactive phenomenon,
as they still maintain clear distances between the audience and the object of contemplation. Under
this sympathetic understanding of interactivity, the work is perceived by the audience as autonomous,
independent, and finished. It requires perspective and distance, therefore eliminating the viewers
39:5
ability to manipulate or change the work of art in concrete ways. This leads to the conclusion that
interactivity can be understood in two ways:
(1) Contemplative artistic reception (reading and reinterpretation)present in traditional films as
shown in the examples above.
(2) Participative artistic reception (cocreation)active involvement of the audience and ultimate goal
of an interactive filmic experience.
Conceiving art in participative and cocreative forms implies a radical suppression of the distances
between artist, work, and audience. It requires restructuring the work of art, planning specific spaces
for audiences intervention and eliminating traditional hierarchies. It is a form of cultural industry,
and its product is a form of mass culture that requires and legitimizes a crowd that, as illustrated by
Canetti [1984], wants to bring everything closer, suppress distances, and eliminate marginalizations.
It is undoubtedly true that interactivity in traditional filmmaking has not been explored to its full
potential, and that with the digitalization of cinema and its transposition to computers and software, it
is now possible to create a radical transformation of interactivity, where users can determine the plot
through actual intervention of the film (participative artistic reception). Regarding this interactive
potential, the focus must be put on the decisions that must be made, how these affect the impediments
that our heroes are confronted with, and how they face these challenges. The impediments within the
story are motivations towards freedom; the story perpetuates itself because the hero chooses. According
to Roland Barthes, within the heros freedom is hidden the survival of the story; in interactive films
the hero is more exposed, and he sacrifices his liberty in favor of the opportunity to include the viewer
in his deliberations and choices. The final result is not only a participative and co-creating viewer, but
also a mutual codependence between the story and its audience.
4.
INTERACTIVE MODELS: FROM A BRANCHING NARRATIVE TO A DETOUR NARRATIVE
One of the main problems with labeling a system as interactive is that the criteria to determine if
it possesses said quality is very broad, vague, and most of all, subjective. Regarding video, one could
argue that traditional playback options (play, pause, forward, rewind) along with random access (like
scene selection in DVDs) are enough to consider a system as interactive. At the same time, a hypothetical interactive drama system where users could freely determine what is to happen next within
the plot is unarguably an interactive system as well. Also, as exposed before, certain interactive behaviors can be identified within traditional films. So, how much interaction does it take to transition
from films to interactive films? There are no definitive answers and, since we are exploring the early
stages of development of interactive films, all approaches towards it are likely to change in time, as
more research and experimentation is done.
The first idea that comes to mind when facing the possibility of interactive films is total freedom.
In other words, we picture unlimited interactivity, where users have absolute autonomy to do as they
like and conduct the plot as they wish. A system that allows such level of personalization obviously
remains in the sci-fi realm and is still far from todays technological reality, but imagining it allows us
to reach a very important conclusion: In such a system we would no longer need a script, therefore we
could no longer tell a specific story. This proves that interactivity is the opposite of narration [Polaine
2005] because if we lose the storytelling ability, it is no longer a narrative act.
Knowing that interaction must be limited in order to maintain the narrative aspects of filmmaking, we have considered two narrative models used in video game storytelling: branching narrative
and string of pearls. The branching narrative model is based on turning points within the script that
branch the game/film into two different stories. In this model, players/viewers chose from pre-designed
narrative paths [Brand and Knight 2005] allowing them to have an effect over the plot, while at the
39:6
R. Verdugo et al.
Fig. 2. Branching narrative.
Fig. 3. String of pearls.
same time allowing writers to tell particular stories. The problem is that the number of different stories grows exponentially (Figure 2), which is the origin of a technical limitation, because the effort and
resources required to write, and later produce such a film would sooner or later become unattainable.
Even if this technical limitation was solved, a bigger, aesthetic limitation would remain because narrative works are usually concerned with immersing the reader or audience in the story and the narrative
suffers when interaction is simply grafted onto it. Either the characters become flat because they are
repetitive, pre-recorded elements triggered by the user or because they are capable of acting in so many
different ways (corresponding to interactive options) that they cease to have character. [Polaine 2005].
Whenever we attempted to write branching narrative scripts we always hit a roadblock: either the
original story had changed so much that it was no longer aesthetically appealing, or the script was
so big that it was no longer a single script, but rather a collection of different scripts with the same
beginning and radically different endings. Filming them would have been the equivalent of filming
many movies at the same time.
After facing this problem we were determined to find a model that would allow us to tell different
stories while at the same time providing us with a structure where branches would not grow exponentially every time viewers interacted with the system, but rather have common paths that would
allow different versions of the story to have shared footage. The string of pearls model used in video
games (Figure 3) considers the story to be composed by a series of pre-set events [Brand and Knight
2005] that are structured linearly (in Figure 3 represented by the arrows), and different worlds that
can be explored (in Figure 3 represented by the pearls or circles). In video games, each of the pearls
is a world, and players are able to move freely inside each of them. But in order to progress in the
story, the player must first successfully perform certain tasks [Handler 2008]. Although the freedom
to explore controlled worlds is desirable in video games, it presents a narrative problem in interactive
filmmaking. As with the total freedom problem described before, each pearl in this model becomes
incontrollable from the writers perspective, and the ability to tell a specific story within them is lost.
Therefore, the problem is that branching narrative allows writers to control the stories being told,
but the tree structure of the scripts grows exponentially every time the story branches out, while on
the other hand, string of pearls allows stories to follow a fixed and controlled path of events, but does
not provide authors with proper control of the narrative process in the different worlds that can be
explored. By mixing both structures it is possible to solve these problems, while keeping the benefits
39:7
Fig. 4. Detour narrative.
that each model provides. We have called this new model, a fusion between branching narrative and
string of pearls, detour narrative. This model has a backbone of common events thatno matter what
story is toldremains unchanged, and is always seen by the audience (similar to the pre-set events
in the string of pearls model). At the end of each of these backbone events viewers can interact with
the system and detour from the events in the backbone to different actions, determined by their input.
Here each pearl, instead of being a freely explorable world, consists of a branching narrative that
branches only once, and therefore eliminates the exponential growth of the tree. After the detour is
over, the system navigates back to the following backbone event. The structure of this detour narrative
model can be seen in Figure 4.
5.
COCONSTRUCTION
5.1 Using Context as a Tool for Coconstruction between Audience and Author
The main benefit of detours is that they can have great effects over the plot, while at the same time
following a similar path of scenes, independent of the branch that is chosen. Instead of basing the different stories on different footage (as in the branching narrative model), we believe that it is possible
to provide different contextualization to scenes (the detours), and that this will lead to different interpretations or readings. Our inspiration comes from the analysis of works of art, and the different ways
they may be perceived, according to the information or context that one has. Consider for example
Vermeers Mistress and Maid (Figure 5). Try to imagine the story behind this very simple scene, where
a woman is interrupted by her maid, carrying a letter. How would that story you just invented change
if you knew the woman was having a secret affair? Or if you knew her husband was overseas at war?
The idea behind this is the distinction between internal and external context [Nack 2003]. External
context corresponds to the information we have about the painting, its environment, the elements depicted in it, and all other structural elements regarding its content, while the internal context enables
39:8
R. Verdugo et al.
c
Fig. 5. Mistress and Maid by Johannes Vermeer. (The
Frick Collection).
the perceiver to emphasize, interpret and evaluate the sources based on the comparison with existing
memory structures (The inner world model). [Nack 2003]. External and internal context are related
to each other through the fact that the internal context will be changed by the external context when
the viewer reacts to the work of art (a painting, a film, etc.). Reading and interpretation will take
elements from the external context and analyze them from the viewers own internal context, which
means that the sense making and final interpretation happen on the viewers side, and cannot be fully
determined by the images provided to the audience. We believe that this process has not been fully explored as a tool for creative and narrative experiences because, as noted by Nack [2003], experiential
systems usually operate in and on the external context whereas the inner context, which forms the
essential aspect of the experience making process, namely the evaluation and instantiation, is hardly
ever modeled.
Our coconstructive model of storytelling bases its narrative on the way we structure the external
context, so that we can influence and control the ways in which the internal context will understand
and make sense of the scenes of the film. By interactively determining a scenes context we can provide
different contextualization to the backbone events within the plot and, through these changes in the
external context of the film, have an effect over the way that the main events are interpreted and read.
This allows us to play with the surroundings of the backbone events without having to change them
directly. Suppose that Vermeers Mistress and Maid was our backbone event, and we wanted to tell two
different stories based on it, without the need of having two different paintings (Figure 6). In one story
the woman has a secret affair (Context A), and in the other her husband is overseas at war (Context
39:9
Fig. 6. Example of coconstruction.
B). These two stories could be told using coconstruction, by offering two different contexts during the
detour that comes after the showing of the painting,. After viewing the painting, not really knowing
what the letter meant or how it affected the story, the audience is given either context, and depending
on what they see their inner context will interpret the scene of the painting differently.
When compared to a branching narrative tree that grows exponentially, it is possible to see that
detour narrative, using coconstruction, allows us to control and diminish the need for shooting additional scenes, by relying on the fact that the meaning and relevance of each scene is not embedded into
the film, but rather interpreted through dialogue between audience and plot. If we look backwards in
time and search among the multiple stories and narrative experiences we have had throughout our
lives, we quickly notice that coconstruction is difficult to apply, and in some cases, impossible. This is
because our culture is lineal; we see everything through the prism of history, where things can only
happen once, and out of all the possible outcomes, just one occurs. In fiction, mutually exclusive events
have been banished because we always commit to one particular option. This illustrates the fact that
our model of interactive films is a cultural construction that requires writers to conceive new stories in
new ways, that break the old mold. We propose actively using audiences as writers that, through interaction, not only determine what scene is to be shown next, but also interpret these scenes differently
according to their own reading.
5.2 Empirical Experimentation with Coconstruction
The main idea behind the concept of coconstruction is the fact that images, scenes, or even entire films
depend not only on the contents depicted, but also on a viewer that reads them and builds a story
with them. In a way, what it ultimately means is that, as narrators or storytellers, we can control only
part of the narrative process, namely, content distribution, but regarding interpretation, we can only
guide audiences into what the meaning behind that content is. Our model of interactive storytelling, by
becoming aware that everything we show to a viewer will ultimately be deconstructed and then reconstructed again, tries to influence the way the audience interprets the scenes that are shown to them.
To have a further understanding of the way co-construction works and how it could be used as a
tool in the script writing of interactive narrative experiences, we developed an experiment to study
the effect that context and interpretation has over the stories that are built from a given content.
We created 4 images (seen in Figure 7) that were used as the content we provided, so that participants could arrange them in any sequence and then tell a story based on them. The images show two
39:10
R. Verdugo et al.
Fig. 7. Pictures of our coconstruction experiment.
Table I. Coconstruction Experiment Results

Main situation
Someone asks another person for a favor
Someone looses something
An object should be inside the black bag, but isnt
Other
Total Stories
Position of the blue image

1st
2nd
3rd
4th
12
1
1
1
3
0
1
1
3
2
1
8
3
2
0
0
21
5
3
10
Total
15
5
14
5
39
characters in different scenarios either using or exchanging a black bag. The black bag appears in three
of the four images, and the objective of the study was to see how the image that does not show the black
bag (the one with the Blue identifier) was differently contextualized and interpreted according to the
position it had in the sequence the participants determined.
5.2.1 Methodology. 48 freshman-year high school students participated in the experiment. Each
student was given the set of 4 pictures seen in Figure 7. Every student received the images in a random
order, and they were identified by the colored dot on the upper-right hand corner of each image. The
random order and color identifiers (instead of letters or numbers) were used so that the images had
no predetermined order. The students were then given sheets of paper with the following instructions:
Arrange the images in any sequence you wish. Write down the sequence and then, based on the
images in the sequence you chose, tell a story explaining what happened. No further instructions
were provided; there was no time or space constraint, and participants were not allowed to talk to each
other during the experiment.
5.2.2 Results. A reviewer read each story and eliminated 9 that were either illogical or departed so
much from what is shown in the pictures, that it made it hard to identify the presence of each image
in the story. The 39 remaining samples where then classified according to the main situation described
in the story, and also according to the position of the image with the blue identifier (Table I).
Every story was reread and the parts that referenced the blue image were identified. Every sentence
or group of sentences was catalogued according to the function it served to the story being told. In 19
of the 21 stories where the blue image is in the first position, it is used to define a situation in which
the black bag will be either eliminated, thrown away, or hidden; in these stories the blue image is
fundamental to understand what will happen to the bag, and why the characters must get rid of it.
In all 5 stories where the blue image is in the second position, it is impossible to distinguish when
the story is referencing the picture that is left in the first position, from when it references the blue
image; all the participants of the experiment that left the blue image in the second position merged
the actions that happened in this picture with whatever image they put in the first position. In 2 of the
3 cases where the blue image is put in the third position, the image is used to explain what was inside
of the black bag and why it is important to recover it. Finally, in 9 of the 10 stories with the blue image
in the last position, the image is used to explain the motivations or the drive that lead the characters
39:11
to get rid of, or recover the black bag. In a way, these stories follow a very similar path to the majority
of those in which the blue image is in the first position except for the fact that, when located in the last
position, the blue image acts as a revelation or final surprise, while when it is used in the first position,
it serves as a preparation of the scenario where the action will unfold.
5.2.3
Analysis. There are three important observations regarding this experiment.
(1) Despite the narrative freedom given to participants, the number of stories was not infinite, but
rather limited. The vast majority of participants told one of three basic stories.
(2) Despite the random order in which the images where handed out and the fact that there was no
identifiable predetermined sequence, the blue image appears -in the majority of cases, in the first
or last position.
(3) According to its position within the sequence, the blue image serves a specific narrative purpose.
Regarding coconstruction, it is possible to enunciate three important conclusions from these experimental observations.
(1) From a given set of images, scenes, or footage, the vast majority of the audience will coconstruct
a limited number of stories. While it is true that different people interpret and read the contents
given to them in different ways, it is also true that the final result is not an infinite array of radically
different stories, but rather a limited number of stories with slight differences between one another.
The implications this has over interactive filmmaking is the fact that we can potentially control and
influence what those stories are and, by means of understanding what audiences see according to
the context given to a particular scene, it is possible to use the same footage within different stories.
(2) Being able to tell two or more stories from a limited set of images or footage does not depend
on radical changes and completely different structures, but rather on slight differences and small
changes in the context in which a certain scene is presented. Despite the fact that the images in our
experiment could be arranged in 24 different sequences, almost 80% of participants used the same
7 sequences. This shows that the different stories we can potentially tell are hidden behind subtle
changes in the way we read actions or events, and not behind complex and intricate structures
that, in cinema, would potentially lead to unattainable footage requirements, along with a much
more complicated narrative model.
(3) If a single image, scene, or footage can serve various narrative purposes depending on its context,
then, instead of basing our interactive films on different rewritings of actions that can serve as
alternative scenes for one another, we must think of ways to repurpose a particular action in
a way that it can tell two different stories based on the same footage; basically, coconstruction
enables us to change the semantics, without having to change the syntax of a scene.
6.
ELEMENTS AND COMPONENTS OF INTERACTIVE FILMS
What changes must be done to the structure of films so that we can allow users to navigate the interconnected paths of a detour narrative? How can we model our Interactive Films, so that they can
be easily written, produced, and then virtually played back, allowing user interaction? How can we
present the film to audiences, allowing them to interact with the film, without having to stop the
movie and ask them what is to happen next? We have divided the answers to these questions into two
groups: Structural Issues and Interface Issues.
6.1 The Structure of the Film
6.1.1 Microcores (MC), Scenes, and Sequences. Within regular films, a scene is considered to be a
succession of shots that happen within the same set, location or space. A sequence can be defined as
39:12
R. Verdugo et al.
Fig. 8. Screen.
a succession of related shots or scenes developing a single subject or phase of the story, forming a
distinct narrative unit. When faced with the necessity to include interactivity within the script, we
realized that none of these structural elements of traditional films would help us define the basic
structural unit of Interactive Films. We called this unit Microcore (MC) and defined it as. The shot, or
series of shots, that contain the lineal fragment of film that shows either a backbone event, or a detour
event within the plot.
According to this definition, the film is composed of two distinct units: Backbone Microcores (BBMC)
and Detour Microcores (DTMC). The distinction between them is that a BBMC links to multiple DTMC,
and the user must navigate to one of them, while DTMC link exclusively to the next BBMC. In terms
of interactivity, BBMC are interactive, while DTMC are not.
Note that the definition purposely uses the term shot, and not scene or sequence, because depending
on how much interaction is put into the script, the following duality can occur: a Micro-Core can contain
a series of sequences or scenes, while alternatively a sequence or scene can contain a series of MicroCores. By allowing flexible relations between traditional film structure and Interactive Film structure,
the model is considerably more expandable and general.
6.1.2 Interactive Moment (IM). Our model considers films to be composed by many Backbone MicroCores that link to multiple Detour Micro-Cores, that then link back to another Backbone Micro-Core.
Between a BBMC and a DTMC there is a navigation point where the users input determines what
DTMC is shown next. For this purpose we have defined an Interactive Moment (IM) as. The place
within the script where it has been stipulated that, during a specified time window, the users input
can be received. It is always located within a Backbone Microcore and is composed by: A decision that
must be made by the viewer, the different options he/she has and finally, the Detour Micro-Core to be
shown next, according to each option.
Detour Micro-Cores do not have Interactive Moments, because once they are over, they navigate to
a particular Backbone Micro-Core and no options are given to the user.
6.2 Display and Interface
6.2.1 Screen. Displaying interactivity is not only a conceptual issue, but also an aesthetic one.
Therefore, we propose a model that easily adapts to different filmic styles, and does not limit artists
full creative potential. We divided the screen into two distinct areas (Figure 8) and created a system of
overlaying objects that are placed on top of the video. The areas are possible action space, and action
space, while dynamic objects and static objects compose the overlaying objects that appear on top of
the screen.
6.2.1.1
Action Space. The action space is the part of the screen where the movie is shown.
39:13
Fig. 9. Action space, possible action space, and a dynamic object.
6.2.1.2 Possible Action Space. The possible action space is a border that surrounds the action space,
where different objects allow users to have an effect over what is shown in the action space.
6.2.1.3 Dynamic Objects. Dynamic objects are images that are overlaid on the screen and can be
dragged and dropped by the user, within both the action space and the possible action space. According
to the position where these are left, logical conditions determine the users semantic action.
6.2.1.4 Static Objects. Static objects are images overlaid on the screen into a fixed position. They
cannot be moved, but they may be used as buttons or drop areas for Dynamic Objects.
6.3 Example
Consider Figure 9, where we can identify the action space (two men talking), and the possible action
space (the black border with the shotgun). Suppose we have reached an Interactive Moment within
this Backbone Micro-Core: the two characters are having a fight, and the one wearing the white shirt
says If I had a gun Id kill you. Immediately afterwards, the shotgun appears underneath the scene
(Figure 9), in the possible action space. During a specified time window, determined by the Interactive
Moment, the shotgun is available as a dynamic object. Users can drag the shotgun anywhere around
the screen, and hand it to the character, or not. After the time window is over, the following Boolean
condition is determined:
If (shotguns final position == action space)
Go To Detour Micro-Core Shooting
Else
Go To Detour Micro-Core Not Shooting
39:14
R. Verdugo et al.
With this logical condition, we are able to determine if the user has decided whether the man should
have the shotgun or not, and what DTMC should be played next. If the user does not move the shotgun,
the final position of the gun is the original one, which means the condition is false, and the object is
not included into the action. This shows that there is a default path, because even if the user does
nothing, a decision has still been made. The main benefit of this particular form of interface is that the
movie never stops to receive the users input. Instead of stopping the movie and asking the viewer Do
you want to give the shotgun to the man in white?, we have done this dynamically. Without losing
immersion or flow, users can interact with the system, while the action continues to unfold.
7.
AN EXAMPLE OF IMPLEMENTATION OF DETOUR NARRATIVE WITH COCONSTRUCTION
In order to test the concepts related to coconstruction and detour narrative, we called on an interdisciplinary team of engineers, filmmakers, visual artists, writers, and producers. We presented them with
the detour narrative model and coconstruction, and began to work with them on our first interactive
short film. Our main goal was to show how interactivity could be achieved, without the need of exponentially growing branching narrative structures, and how certain Backbone Micro-Cores could be
differently interpreted by allowing viewers to change the context of certain scenes. The interactive film
that resulted from this experience is available at: http://cineinteractivo.ing.puc.cl.
7.1 Fernando Morenos Crime or Revenge: An Interactive Short Film
Fernando Morenos Crime or Revenge is the story of a man, Fernando Moreno, who travels to an isolated
countryside. With a rifle on the backseat of his old Chevy pickup, he drives fast and determined. As
clear as the thick track of dust he raises behind his path, Fernando has one goal on his mind: to find
the man he is looking for, and kill him. He might be a justice enforcer, or a hired assassin; a family man
forced to commit a crime, or a merciless avenger. Fernando Morenos reasons to kill, his intentions, and
his fate at the end of the story, are in the audiences hands, and they will be the result of the interactive
decisions they make during the film and the way they coconstruct the story beyond the script.
7.2 Constructing the Film: Considering Audiovisual and Script Issues
We began by establishing the title of the film: Fernando Morenos Crime or Revenge. The title acts
as a domain definition; we are telling two different stories and, thanks to interactive participation,
the audience can navigate either one of them. If this were a traditional film, it would be called either
Fernando Morenos Crime or Fernando Morenos Revenge, depending on the story being told. Because of
interactivity, and the fact that both stories are present, we decided to have an interactive title as well.
Because of subtle differences between the detour events contextualizing the backbone events, each
story being told has multiple versions, with different shades of meaning. The Backbone and Detour
Micro-Core structure of the film is the one shown in Figure 10. There are three BBMC, six DTMC, and
five Dynamic Objects used for navigating the films structure. In Figure 10, lines between Micro-Cores
indicate the different paths that can be followed, and the drawings indicate the objects that must be
dragged into the Action Space from the Possible Action Space for that path to be taken.
The way the script was written provides further insight on how we arrived at this structure for the
film. The first thing that was determined was the main chain of events that would compose the plot;
first, a character (Fernando Moreno) would be seen practicing his shooting skills; second, Fernando
Moreno would be looking for someone, and would receive indications on where to find that person;
third, Fernando Moreno would follow the indications he received and shoot a man from a distance.
These actions became the Backbone Micro-Cores of the film, and no matter what choices are made by
the audience, they are always shown.
39:15
Fig. 10. Micro-Core structure of the film and dynamic objects.
Once the main sequence of events was determined, we began to think of different options that would
link these events with one another. The idea was to connect them with detours that would allow users
to co-construct different interpretations for the Backbone Micro-Cores. Coconstruction was used in two
different, yet complimentary ways. One form of coconstruction occurs when audiences build Fernando
Morenos psychological profile and personality. By understanding his motivations for killing and the
reasons why he is there, audiences are able to build the character from their own point of view. Every particular combination of Micro-Cores has its own profile, which viewers can read and interpret
differently. During the first Interactive Moment (in BBMC 1), a map is available in the possible action space. When dragged into the action space, DTMC A is shown next, and viewers see Fernando
Moreno asking a passing girl for directions. If the map is not dragged into the action space, DTMC B
39:16
R. Verdugo et al.
is shown next, Fernando Moreno is seen talking to a girl, and the dialogue between them reveals that
they already knew each other. The purpose of this interaction is for viewers to coconstruct their own
reading regarding Fernando Morenos connection to the place where the action occurs. The second Interactive Moment (in BBMC 2) allows viewers to receive further information about Fernando Moreno.
Audiences can choose between a shotgun and a cell phone. If the cell phone is dragged into the action
space, DTMC D is shown, and Fernando Moreno is seen talking to his son over the phone while, if
the shotgun is dragged into the action space, DTMC C is shown and Fernando Moreno is seen coldly
practicing his shooting skills. Detours, in this case, enable audiences to coconstruct Fernando Morenos
state of mind right before the killing occurs. Additionally, as seen in Figure 10, the structure of the film
allows users to include neither the shotgun nor the cell phone into the action space, in which case detours are skipped and the following BBMC is shown. This was done to experiment with a more flexible
structure and to allow more open and ambiguous constructions of the film, that have more space for
interpretation and coconstruction.
The second form of coconstruction occurs in the reading of the final Backbone Micro-Core. Fernando
Moreno kills a man, and it is up to the viewer to interpret what has happened. The third Interactive
Moment (in BBMC 3) determines two alternative endings. In one case, Fernando, after killing a man,
visits the shrine of Alejandra Moreno and leaves the bullets as a sign that justice has been done (DTMC
E). In the other case, he talks to his boss over the phone, and goes back to the place of the killing
(DTMC F). Who is the man killed by Fernando Moreno? Has he killed the wrong man? Is he a hired
assassin, or an avenger? Although the killing that takes place in BBMC 3 is the same, no matter what
version of the movie is seen, the different contexts to the killing allow viewers to come up with their
own answers to these questions and, through their coconstructive involvement, the final story is built
by them, instead of being fully determined by the script.
7.3 Writing, Filming, and Programming
Once the films structure was determined, over a series of brainstorming sessions that took place during a two-week period, the final script was written. Each scene was written independently, and then
all possible combinations of the script were handed out to the team, to check for consistency and flow
of the story. During this period, some minor modifications to the scenes were done. Filming took place
during three days, after a two-week preproduction period. Each person in the acting and production
crew had a printed version of the films structure, and the detours were shot as alternative scenes.
Every time a scene was shot, the script supervisor checked for continuity issues across all possible
navigation paths. Afterwards, during a three-week period, each micro-core was edited as an independent chunk of film, using regular video editing software. By creating playlists with the independent
chunks of film, each navigational path was seen as an independent film. Once all micro-cores were
rendered as independent movie files, and each dynamic object was illustrated, the interactive version
of the film was created, over a three-day period, using Flash. In the structure of the software, each
micro-core is inserted into a different key-frame of the flash movie, using an FLVPlayback component.
This allows us to separate the films structure and playback software from the actual chunks of video
that compose the film. This flexible structure easily enables the possibility to change the films microcores, without having to change the source code (for example, the version with English subtitles uses
the same software, but different videos than the version without them). The timing of the dynamic
objects fade-in and out, and the navigation between one micro-core and the next, is done by embedded
cue points that act as metadata for each video chunk. Boolean conditions, according to the dynamic
objects final position within the frame, evaluate the audiences input. Overall, the softwares structure
is very simple and modular; therefore it is easy to create new films reusing the code written for each
39:17
key-frame. Changing the navigational paths of the film is also very simple, because it only involves
changing the Boolean conditions evaluated at the end of each video.
7.4 A First Encounter with Interactive Filmmaking: Fernando Morenos Crime or Revenge
in Front of an Audience
We invited 42 students from the Film Program at our Universitys School of Communications, to individually experience Fernando Morenos Crime or Revenge at one of our computer labs, and then
answer a survey that collected their opinion regarding the film, the system, and interactive filmmaking
as a concept. The study was conducted with film students, and not general audiences, because in order
for interactive films to exist in the future, we first need innovators and early adaptors who embrace the
concept and create interactive audiovisual content that can later be distributed, and commercialized.
7.4.1 The Study. Each of the 42 students who saw the film at our lab took a survey divided into
three sections. Each section had a particular objective.
(1) Section one was aimed at assessing coconstruction in the narrative of the film.
(2) Section two measured usability, and user experience of the interactive system.
(3) Section three collected the users opinions about the future of interactive filmmaking.
7.4.2 Assessing Coconstruction. To measure the impact of co-construction, each participant was
asked to watch the movie while the interactive options they took were registered. Then they were
asked to answer the following open-ended questions: Why has Fernando Moreno come to this town?
and Who was the person killed by Fernando Moreno?
What we were trying to determine was whether or not there is a correlation between the objects
inserted into the film and the reasons that viewers believe motivated Fernando Moreno to kill a man.
We chose to ask the reasons why he had come to town (instead of directly asking for the reasons of the
killing), and use an open-ended question, so that the question itself would not condition or influence
the answer.
Each answer to the first question was then, standardized according to four categories.
(a)
(b)
(c)
(d)
Fernando Moreno has come to kill a man.

Fernando Moreno has come to kill a man because he is seeking for revenge.
Fernando Moreno has come to kill a man because he is a hired assassin.
Other
Users who were classified into the Fernando Moreno has come to kill a man category did not
specify the information we were looking for, and those who where classified into the Other category
gave elaborate answers that had no reference to the killing itself. After the answers were standardized,
they were classified according to the object the user had chosen during the third Interactive Moment
(Table II).
Among the users that incorporated the Shrine during the third interactive moment of the film,
63.6% explicitly enunciated that Fernando Moreno had come to kill a man because he was looking for
revenge, while none of them mentioned that he was a hired assassin. Among the users that incorporated the Old Man during the third Interactive Moment, 40% explicitly said that Fernando Moreno
was a hired assassin, while 15% (3 users) said that he had come looking for revenge. Because of the
fact that the third Interactive Moment determines the outcome of the story, it can be clearly seen that
the majority of users who chose the Shrine saw Fernando Moreno as a man looking for revenge, while
the majority of those who chose the Old Man saw Fernando Moreno as a hired assassin. No users identified Fernando Moreno as a hired assassin when they chose the Shrine, but three identified him as a
39:18
R. Verdugo et al.
Table II. Why Has Fernando Moreno Come to Town?
Third Interactive
Moment
Kill a Man
Revenge
Hired Assassin
Other
14
Shrine
Old Man
Table III. Who is The Man Killed by Fernando Moreno?

Third Interactive
Moment
Not the man he

was looking for
Undetermined
The man he was looking

for (Roberto Daz)
20
17
Shrine
Old Man
man looking for revenge, despite having chosen the Old Man. To look for an explanation for this, we
checked the other options these three particular users had taken during the first and second Interactive Moments, and found that all three of them had incorporated the Map during the first Interactive
Moment. Because of this, during the first Detour they saw the conversation between the young girl and
Fernando Moreno asking for directions. During that conversation Fernando says, Im here to collect
some debts which can be interpreted, in Spanish, as Im here to get even with someone. Those three
users may have interpreted that scene differently because of that dialogue, or because they may have
been influenced by the title of the film.
The second open-ended question, Who is the man killed by Fernando Moreno? provides information regarding the way coconstruction allows us to rewrite the meaning of certain events, and how
viewers interpretation can be influenced by the context provided to the backbone events of the plot.
What we wanted to understand was how, according to the interactive preferences incorporated by the
viewer, the scene where Fernando Moreno killed a man could be read in different ways. Answers were
classified according to whether or not viewers had mentioned that Fernando had killed the wrong person, or if he had killed the man he was looking for (Roberto Daz), and also according to the object
chosen during the third Interactive Moment (Table III).
39:19
Table IV. Footage Requirements

Micro-Core Group
Backbone
Detour A
Detour B
Duration
4:09
5:13
5:19
Percent of Total
28,2%
35,5%
36,2%
Because 90% of the viewers who incorporated the Shrine considered that Fernando Moreno had
killed the right man, while 85% of the viewers that incorporated the Old Man mentioned that he had
not killed the man he was looking for, it can be clearly seen that, according to the context determined
by the third Interactive Moment, the final Detour rewrites what audiences interpret in the scene of
the killing.
In Table IV we show a brief analysis of the overhead in footage requirements introduced by interactivity compared with a lineal film. To do this we grouped the Detour Micro-Cores into two different sets
(A and B). Set A considers the incorporation of the Map during the first Interactive Moment, the Cell
Phone during the second Interactive Moment and the Old Man during the third one. Set B considers
all other actions. If Fernando Morenos Crime or Revenge was a lineal film, only one set of Detour
Events would have been necessary, therefore, the other set can be considered additional requirements
in footage because of interactivity. With an average run time of 9 minutes and 25 seconds, and a total of
14 minutes and 41 seconds of edited footage, our model required approximately 5 minutes of additional
footage to make interactivity possible.
The final question in Section One of the study allowed participants in the experiment to leave comments and feedback regarding the film. Three users commented that they were frustrated when, despite not having chosen the Shotgun during the second Interactive Moment, it appeared anyway
during the killing. This is a problem that was not foreseen during the writing of the script. The detour after the second Interactive Moment uses the shotgun as a form of configuring Fernando Morenos
character, by showing him as a cold blooded criminal instead of what happens in the alternative detour,
where he is shown as a father calling his son. We did not realize that the shotgun would appear anyway
at the following backbone event. A solution for this is to eliminate the Shotgun as an option during
the second Interactive Moment, leaving only the Cell Phone as a dynamic object, and using the other
scene as a default path when the Cell Phone is not dragged into the action space. A new version of
the film with this structure is available online at http://cineinteractivo.ing.puc.cl/newversion.
7.4.3 User Experience and Usability. Section 2 of the study evaluated the systems interface and
assessed the overall user experience. Users level of agreement or disagreement to 20 different statements was measured using a five-level Likert scale. Table IV shows the summary of the results.
All items measuring the systems usability reflect that the vast majority of users evaluate it positively, and consider the system as simple (items 1 and 18), intuitive (item 7), easy to use (item 11 and
12), easy to remember (item 13 and 14), and responsive (item 19). Items regarding errors and exceptions show that most users did not experience any major difficulties (items 5, 15 and 16) although in
those same items it is possible to identify a single user who evaluated the system poorly due to a computer crash. Finally, regarding the overall experience, a remarkably high percentage of users evaluate
the experience as fun and agree that they felt comfortable interacting with the system (items 4 and 17).
7.4.4 Evaluating Interactive Cinema. Because the participants in the study were all film students,
we were interested in their opinion regarding Interactive Cinema in general, and not just this particular short film. Section 3 of the study evaluated participants opinion regarding the broader idea of
Interactive Cinema, and the model we have developed. Users level of agreement or disagreement to 7
different statements was measured using a five-level Likert scale. Table V shows the summary of the
results.
39:20
R. Verdugo et al.
Table V. Usability and User Experience
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Icons that represent the objects are clear and simple

to understand.
The objects inserted into the film had an effect over
the plot.
I wish there were more moments to incorporate
objects into the film.
I felt comfortable interacting with the system.
I had problems incorporating objects into the film, or
interacting with the system
The elements/objects I added to the movie had an
important role in the unfolding of the action.
The system is intuitive and easy to use.
Because of interactivity and having to use the system,
I sometimes got distracted and stopped paying
attention to the unfolding of the story.
Initial instructions were sufficient to understand how
to use the system.
I never understood what the objective of interacting
with the system was.
It was easy to know what to do when I had the option
to interact with the system.
I had to learn many things to use the system.
Once I learned how to use the system, I had no
problems interacting with it.
If I saw an Interactive Film a month from today, I
would remember how to use the system.
There were situations in which I tried to perform an
action, but wasnt able to.
I had to stop watching the movie because of errors in
the system.
The Interactive Film was fun to watch.
The system is simple.
The system is quick and responsive.
I think that the system lacks some features.
Strongly
Agree
64,3%
Agree
33,3%
Neither Agree
nor Disagree
0%
Disagree
2,4%
Strongly
Disagree
0%
36,6%
48,8%
9,8%
4,8%
0%
59,5%
28,6%
9,5%
2,4%
0%
47,6%
2,4%
33,3%
4,8%
14,3%
7,1%
0%
14,3%
4,8%
71,4%
11,9%
47,7%
19%
19%
2,4%
66,7%
0%
26,2%
14,2%
4,8%
26,2%
2,3%
28,6%
0%
31%
73,8%
21,4%
2,4%
2,4%
0%
0%
2,4%
14,3%
28,6%
54,7%
54,8%
35,7%
7,1%
2,4%
0%
0%
85,7%
2,4%
14,3%
0%
0%
21,4%
0%
76,2%
0%
83,3%
14,3%
0%
2,4%
0%
2,4%
11,9%
9,5%
26,2%
50%
2,4%
0%
4,8%
19%
73,8%
47,6%
81%
71,4%
5,5%
38,1%
19%
26,2%
16,7%
14,3%
0%
0%
16,7%
0%
0%
2,4%
25%
0%
0%
0%
36,1%
Statements included in the study evaluated general opinion regarding Interactive Cinema (Items 4
and 5), as well as personal preferences (Items 1, 6, and 7). Results show a very high level of approval
regarding Interactive Cinemas future as a viable form of entertainment, and slightly lower ratings
when it comes to personal preferences. This shows that the majority of future filmmakers who participated in the study recognize the value of Interactive Cinema, even if they dont prefer it themselves.
Finally, items 2 and 3 show the openness and eagerness of future filmmakers to experiment with new
ways of distribution of audiovisual content, like the Internet and mobile devices.
8.
CONCLUSIONS
This article reflects our approach towards Interactive Filmmaking, and shows that this vast and innovative media can provide us with creative and unique tools for exploring the hidden narrative worlds
behind a what would have happened if. . .. Interactive films, despite the new challenges and problems
they propose, have the potential to become a powerful and refreshing reinvention of a media born
through technological innovation, which has always been open to change. In Mark Cousinss [2004]
39:21
Table VI. Evaluating Interactive Cinema
1
2
3
4
5
6
7
I would like to see more Interactive Films in the future.

I would like to see Interactive Films on my cell phone
or other mobile devices.
I would like to see Interactive Films on the Internet
I believe that Interactive Cinema has a future in the
industry of entertainment
I believe that interactive experiences are the future of
communications and entertainment.
Experiences like Interactive Cinema are the type of
entertainment I would like to enjoy in the future.
I would like to see an Interactive Feature Film.
Strongly
Agree
55,3%
26,3%
Agree
36,8%
23,7%
Neither Agree
nor Disagree
7,9%
39,4%
Disagree
0%
5,3%
Strongly
Disagree
0%
5,3%
57,9%
42,1%
31,6%
52,7%
7,9%
2,6%
0%
0%
2,6%
2,6%
36,8%
36,8%
18,5%
7,9%
0%
35,9%
38,5%
12,7%
10,3%
2,6%
63,2%
26,3%
5,3%
2,6%
2,6%
words, It is helpful to imagine cinema evolving as a language or replicating like genes because doing
so illustrates that film has a grammar and that in some ways it grows and mutates.
Fernando Morenos Crime or Revenge is a first approach towards Interactive Filmmaking using detour narrative and coconstruction. In the study we conducted with film students, both variables that
were indirectly measuredif Fernando Moreno killed the right man or not and if he is a hired assassin
or an avengershow that our model for interactive filmmaking allows writers to use audience interpretation as a tool for building audiovisual experiences that narrate two (or more) different stories.
Viewer reception and interpretation of a film is not new. Coconstruction, which is the intentional use
of this reception and interpretation as a narrative tool within an interactive environment, is.
We believe that the detour narrative model along with coconstruction can be an extremely powerful
tool for creating steerable narrative experiences, and that its full potential is yet to be discovered.
Further empirical investigation must be done regarding coconstruction and its narrative uses. The
following questions remain unanswered and require future narrative experiments, with both audiences
and authors, to shed light on them.
(1) How does the use of coconstruction alter they way we design, write and film audiovisual experiences?
(2) How can authors easily include coconstruction into their scripts?
(3) How does the audiences cultural background influence coconstruction? Is this predictable?
Additionally, it is possible to detect computational requirements within our work that would make
the creation of these audiovisual experiences much easier. Examples of this are the following.
(1) Software based on the coconstructive detour narrative model that allows the production of Interactive Films without the need of programming, following similar graphic structures as the ones used
to produce detail-on-demand video by the Hyper-Hitchcock editor [Shipman et al. 2008].
(2) Nonlinear video editing software that allows easy management of each microcore as a separate
chunk of video.
(3) Production management software and script writing software for nonlinear and branching scripts.
The applications that this narrative model might have in multiple fields like education, interactive
television, e-learning, marketing, art, performance, and others must be explored, and further research
must be done. We hope that this article serves as a precedent for future work in diverse fields and
disciplines, because it is only through interdisciplinary efforts that we will gain further understanding
and insight of the challenges and opportunities that new media provides us with.
39:22
R. Verdugo et al.
APPENDIX
39:23
ACKNOWLEDGMENTS
We would like to acknowledge The Frick Collection for allowing us to use Vermeers painting Mistress
and Maid.
REFERENCES
ATKINSON, S. 2008. Crossed Lines. In Proceedings of the 3rd International Conference on Digital Interactive Media in Entertainment and Arts. 537538.
BRAND, J. AND KNIGHT, S. 2005. The narrative and Ludic nexus in computer games: Diverse worlds II. In Proceedings of the
DiGRA Conference: Changing Views Worlds in Play.
BROOKS, K. 1996. Do story agents use rocking chairs? The theory and implementation of one model for computational narrative.
In Proceedings of the 4th ACM International Conference on Multimedia. 317328.
CANETTI, E. 1984. Crowds and Power. Farrar, Straus and Giroux, New York, NY.
CSIKSZENTMIHALYI, M. 1990. Flow: The Psychology of Optimal Experience. Harper & Row, New York, NY.
DAVENPORT, G. 2002. Art and other anxieties: Steerable stories. In Proceedings of the Conference on Story Telling in the Digital
Age.
DORFLES, G. 1984. El Intervalo Perdido. Lumen, Barcelona, Spain.
ECO, U. 1989. The Open Work. Harvard University Press, Cambridge, MA.
HALLIWELL, L. 1988. Halliwells Filmgoers and Video Viewers Companion. Harper Collins, New York, NY.
HANDLER, C. 2008. Digital Storytelling, Second Edition: A Creators Gude To Interactive Entertainment. Focal Press. Burlington,
MA.
JAUSS, R. 1982. Toward an Aesthetic of Reception. University of Minnesota Press, Minneapolis, MN.
JOHNSON, K. 2008. Lost cause, an interactive film proyect. In Proceedings of the 16th ACM International Conference on Multimedia. Canada. 569578.
KNOLLER, N. 2004. InterFace portraits: communicative-expressive interaction with characters mind. In Proceedings of the 1st
ACM Workshop on Story Representation, Mechanism and Context. 6366.
LEW, M. 2004. Live cinema: an instrument for cinema editing as a live performance. In Proceedings of the International Conference on Computer Graphics and Interactive Techniques. 117.
MACFADYEN, A., STRANIERI, A., AND YEARWOOD, J. 2007. In Proceedings of the 4th Australasian Conference on Interactive Entertainment. Article 13.
MARQUEZ
, M., GONZALEZ
, P., AND OSAWA, K. 2007. Mutable cinema, a participatory narrative engine. In Proceedings of the 2nd
International Conference on Digital Interactive Media in Entertainment and Arts. 181182.
DOMIKE, S, MATEAS, M., AND VANOUSE P. 2002. The recombinant history apparatus presents: Terminal Time. In Mateas, M. and
Sengers, P. Eds., Narrative Intelligence. John Benjamins, Amsterdam.
NACK, F. 2003. Capturing experienceA matter of contextualising events. In Proceedings of the Multimedia Workshop and
Experiential Telepresence (ETP03). 5364.
PERNIOLA, M. 2001. La Estetica del Siglo XX. La balsa de la medusa, Madrid, Spain.
POLAINE, A. 2005. The flow principle in interactivity. In Proceedings of the 2nd Australasian Conference on Interactive Entertainment. 151158.
SAWHNEY, N., BALCOM, D., AND SMITH, I. 1996. HyperCafe: Narrative and aesthetic properties of hypervideo. In Proceedings of the
7th ACM Conference on Hypertext. 110.
SCHIPMAN, F., GIRGENSOHN, A., AND WILCOX, L. 2008. Authoring, viewing and generating hypervideo: An overview of hyperhitchcock. ACM Trans. Multimedia Comput. Comm. Appl. 5, 2.
STENZLER, M. AND ECKERT, R. 1996. Interactive video. SIGCHI Bull. 28, 2, 7681.
SZILLAS, N. 2005. The future of interactive drama. In Proceedings of the 2nd Australasian Conference on Interactive Entertainment. 139199.
TOKUHISA, S., DING, A., AND INAKAGE, M. 2005. Tri-Story as intuitive cinema interactive storytelling based on physical action for multi-screen. In Proceedings of the ACM SIGCHI International Conference on Advances in Computer Entertainment
Technology. 314317.
VATTIMO, G. 1997. Beyond Interpretation: The Meaning of Hermeneutics for Philosophy. Stanford University Press, Palo Alto, CA.
XU, R. Y. D., WANG, R., PARAMESWARAN, N., AND JIN, J. S. 2003. Media streaming synchronization and video interaction: A survey.
In Selected Papers from the Pan-Sydney Workshop on Visualisation. 22. 113115.
39:24
R. Verdugo et al.
FILM REFERENCES
A Bout de Souffle (1960) by Jean-Luc Godard http://www.imdb.com/title/tt0053472/
Funny Games (1997) by Michael Haneke http://www.imdb.com/title/tt0119167/
Lady In The Lake (1947) by Robert Mongomery http://www.imdb.com/title/tt0039545/
Pillow Book (1996) by Peter Greenaway http://www.imdb.com/title/tt0114134/
Rashomon (1950) by Akira Kurosawa. http://www.imdb.com/title/tt0042876/
Time Code (2000) by Mike Figgis http://www.imdb.com/title/tt0220100/
Received October 2009; revised January 2010; accepted March 2010

Interactive Films and Co-Construction

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Interactive Films and Co-Construction

Загружено:

Авторское право:

Доступные форматы

Interactive Films and Coconstruction

RENATO VERDUGO, MIGUEL NUSSBAUM, PABLO CORRO, PABLO NUNEZ,

Verdugo, R., Nussbaum, M., Corro, P., Nunez,

Catolica de Chile; email: pcorro@uc.cl; P. Nunez,

Fig. 1. Issues regarding interactive films.